MPLS WAN Technology Design Guide December 2013
Table of Contents Preface...1 CVD Navigator...2 Use Cases... 2 Scope... 2 Proficiency... 2 Introduction...3 Related Reading... 3 Technology Use Cases... 3 Use Case: Site-to-Site Communications Using MPLS L3VPN Services... 3 Design Overview... 4 WAN Design... 4 MPLS WAN Transport... 4 Ethernet WAN... 4 WAN-Aggregation Designs... 5 MPLS Static Design Model... 6 MPLS Dynamic Design Model... 6 Dual MPLS Design Model... 7 WAN Remote-Site Designs... 7 WAN/LAN Interconnection... 8 WAN Remote Sites LAN Topology... 9 Layer 2 Access... 9 Distribution and Access Layer...11 IP Multicast...13 Quality of Service...13 Deploying the WAN...15 Overall WAN Architecture Design Goals... 15 IP Routing... 15 LAN Access... 15 High Availability... 15 Path Selection Preferences... 15 Quality of Service (QoS)... 16 Design Parameters... 16 Table of Contents
Deploying an MPLS WAN...17 Design Overview...17 WAN-Aggregation MPLS CE Routers...17 Remote Sites MPLS CE Router Selection... 18 Design Details...19 Deployment Details... 22 Configuring the MPLS CE Router... 22 Configuring the Remote-Site MPLS CE Router... 33 Adding a Secondary MPLS Link on an Existing MPLS CE Router... 54 Configuring the Secondary Remote-Site Router... 59 Deploying a WAN Remote-Site Distribution Layer...75 Deployment Details... 75 Connecting the Single or Primary Remote-Site Router to the Distribution Layer... 75 Connecting the Secondary Remote-Site Router to the Distribution Layer... 81 Deploying WAN Quality of Service...86 Deployment Details... 86 Configuring QoS... 86 Appendix A: Product List...91 Appendix B: Device Configuration Files...94 Appendix C: Changes...95 Table of Contents
Preface Cisco Validated Designs (CVDs) provide the foundation for systems design based on common use cases or current engineering system priorities. They incorporate a broad set of technologies, features, and applications to address customer needs. Cisco engineers have comprehensively tested and documented each CVD in order to ensure faster, more reliable, and fully predictable deployment. CVDs include two guide types that provide tested and validated design and deployment details: Technology design guides provide deployment details, information about validated products and software, and best practices for specific types of technology. Solution design guides integrate or reference existing CVDs, but also include product features and functionality across Cisco products and may include information about third-party integration. Both CVD types provide a tested starting point for Cisco partners or customers to begin designing and deploying systems using their own setup and configuration. How to Read Commands Many CVD guides tell you how to use a command-line interface (CLI) to configure network devices. This section describes the conventions used to specify commands that you must enter. Commands to enter at a CLI appear as follows: configure terminal Commands that specify a value for a variable appear as follows: ntp server 10.10.48.17 Commands with variables that you must define appear as follows: class-map [highest class name] Commands at a CLI or script prompt appear as follows: Router# enable Long commands that line wrap are underlined. Enter them as one command: police rate 10000 pps burst 10000 packets conform-action set-discard-classtransmit 48 exceed-action transmit Noteworthy parts of system output or device configuration files appear highlighted, as follows: interface Vlan64 ip address 10.5.204.5 255.255.255.0 Comments and Questions If you would like to comment on a guide or ask questions, please use the feedback form. For the most recent CVD guides, see the following site: http://www.cisco.com/go/cvd/wan Preface December 2013 1
CVD Navigator The CVD Navigator helps you determine the applicability of this guide by summarizing its key elements: the use cases, the scope or breadth of the technology covered, the proficiency or experience recommended, and CVDs related to this guide. This section is a quick reference only. For more details, see the Introduction. Use Cases This guide addresses the following technology use cases: Site-to-Site Communications Using MPLS L3VPN Services Many organizations are deploying Multiprotocol Label Switching (MPLS) WAN services in order to connect remote locations over private cloud Layer 3 VPN-based providermanaged MPLS networks. Related CVD Guides VALIDATED DESIGN Campus Wired LAN Technology Design Guide For more information, see the Use Cases section in this guide. Scope This guide covers the following areas of technology and products: WAN design using Layer 3 MPLS services for central and remote sites Remote-site WAN redundancy options Routing policy and control for WAN aggregation and remote sites WAN quality of service (QoS) design and configuration For more information, see the Design Overview section in this guide. Proficiency This guide is for people with the following technical proficiencies or equivalent experience: CCNP Routing and Switching 3 to 5 years planning, implementing, verifying, and troubleshooting local and widearea networks VALIDATED DESIGN VALIDATED DESIGN GET VPN Technology Design Guide VPN WAN Technology Design Guide To view the related CVD guides, click the titles or visit the following site: http://www.cisco.com/go/cvd/wan CVD Navigator December 2013 2
Introduction The MPLS WAN Technology Design Guide, provides flexible guidance and configuration for Multiprotocol Label Switching (MPLS) transport. Related Reading The Layer 2 WAN Technology Design Guide provides guidance and configuration for a VPLS or Metro Ethernet transport. The VPN WAN Technology Design Guide provides guidance and configuration for broadband or Internet transport in a both a primary or backup role. Technology Use Cases For remote-site users to effectively support the business, organizations require that the WAN provide sufficient performance and reliability. Although most of the applications and services that the remote-site worker uses are centrally located, the WAN design must provide a common resource access experience to the workforce, regardless of location. To control operational costs, the WAN must support the convergence of voice, video, and data transport onto a single, centrally managed infrastructure. As organizations move into multinational or global business markets, they require a flexible network design that allows for country-specific access requirements and controls complexity. The ubiquity of carrier-provided MPLS networks makes it a required consideration for an organization building a WAN. To reduce the time needed to deploy new technologies that support emerging business applications and communications, the WAN architecture requires a flexible design. The ability to easily scale bandwidth or to add additional sites or resilient links makes MPLS an effective WAN transport for growing organizations. Use Case: Site-to-Site Communications Using MPLS L3VPN Services This guide helps organizations deploy WAN services in order to connect remote locations over private cloud Layer 3 VPN-based provider managed MPLS services. This design guide enables the following network capabilities: IP any-to-any WAN connectivity for up to 500 remote sites and one or two central hub site locations Deployment of single or dual MPLS service providers for resiliency using single or dual routers in remote site locations Static routing or dynamic BGP peering with the MPLS service provider for site-to-site communications. Support for Layer 2 or Layer 3 distribution switching designs Support for IP multicast using Multicast VPN (mvpn) service provider-based offering QoS for WAN traffic such as Voice over IP (VoIP) and business critical applications Introduction December 2013 3
Design Overview This guide, the MPLS WAN Technology Design Guide, provides a design that enables highly available, secure, and optimized connectivity for multiple remote-site LANs. The WAN is the networking infrastructure that provides an IP-based interconnection between remote sites that are separated by large geographic distances. This document shows you how to deploy the network foundation and services to enable the following: WAN Design MPLS WAN connectivity for up to 500 remote sites Primary and secondary links to provide redundant topology options for resiliency Wired LAN access at all remote sites The primary focus of the design is to allow usage of the following commonly deployed WAN transports: Multiprotocol Label Switching (MPLS) Layer 3 VPN (primary) Multiprotocol Label Switching (MPLS) Layer 3 VPN (secondary) Internet VPN (secondary) At a high level, the WAN is an IP network, and these transports can be easily integrated to the design. The chosen architecture designates a primary WAN-aggregation site that is analogous to the hub site in a traditional hub-and-spoke design. This site has direct connections to both WAN transports and high-speed connections to the selected service providers. In addition, the site uses network equipment scaled for high performance and redundancy. The primary WAN-aggregation site is coresident with the data center and usually the primary campus or LAN as well. The usage of an Internet VPN transport to provide a redundant topology option for resiliency is covered in the VPN WAN Technology Design Guide. MPLS WAN Transport Cisco IOS Software Multiprotocol Label Switching (MPLS) enables enterprises and service providers to build next-generation, intelligent networks that deliver a wide variety of advanced, value-added services over a single infrastructure. You can integrate this economical solution seamlessly over any existing infrastructure, such as IP, Frame Relay, ATM, or Ethernet. MPLS Layer 3 VPNs use a peer-to-peer VPN Model that leverages the Border Gateway Protocol (BGP) to distribute VPN-related information. This peer-to-peer model allows enterprise subscribers to outsource routing information to service providers, which can result in significant cost savings and a reduction in operational complexity for enterprises. Subscribers who need to transport IP multicast traffic can enable Multicast VPNs (MVPNs). The WAN leverages MPLS VPN as a primary WAN transport or as a backup WAN transport (to an alternate MPLS VPN primary). Ethernet WAN Both of the WAN transports mentioned previously use Ethernet as a standard media type. Ethernet is becoming a dominant carrier handoff in many markets and it is relevant to include Ethernet as the primary media in the tested architectures. Much of the discussion in this guide can also be applied to non-ethernet media (such as T1/E1, DS-3, OC-3, and so on), but they are not explicitly discussed. Introduction December 2013 4
WAN-Aggregation Designs The WAN-aggregation (hub) designs include two or more WAN edge routers. When WAN edge routers are referred to in the context of the connection to a carrier or service provider, they are typically known as customer edge (CE) routers. All of the WAN edge routers connect into a distribution layer. The WAN transport options include MPLS VPN used as a primary or secondary transport. Each transport connects to a dedicated CE router. A similar method of connection and configuration is used for both. This design guide documents multiple WAN-aggregation design models that are statically or dynamically routed with either single or dual MPLS carriers. The primary differences between the various designs are the usage of routing protocols and the overall scale of the architecture. For each design model, you can select several router platforms with differing levels of performance and resiliency capabilities. Each of the design models is shown with LAN connections into either a collapsed core/distribution layer or a dedicated WAN distribution layer. There are no functional differences between these two methods from the WAN-aggregation perspective. In all of the WAN-aggregation designs, tasks such as IP route summarization are performed at the distribution layer. There are other various devices supporting WAN edge services, and these devices should also connect into the distribution layer. Each MPLS carrier terminates to a dedicated WAN router with a primary goal of eliminating any single points of failure. A single VPN hub router is used across both designs. The various design models are contrasted in the following table. Table 1 - WAN-aggregation design models MPLS Static MPLS Dynamic Dual MPLS Remote sites Up to 50 Up to 100 Up to 500 WAN links Single Single Dual Edge routers Single Single Dual WAN routing protocol None (static) BGP (dynamic) BGP (dynamic) Transport 1 MPLS VPN A MPLS VPN A MPLS VPN A Transport 2 MPLS VPN B The characteristics of each design are discussed in the following sections. Introduction December 2013 5
MPLS Static Design Model Supports up to 50 remote sites Has a single MPLS VPN carrier Uses static routing with MPLS VPN carrier The MPLS Static design model is shown in the following figure. Figure 1 - MPLS Static and MPLS Dynamic design models (single MPLS carrier) Core Layer Collapsed Core/ Distribution Layer Distribution Layer MPLS CE Router MPLS CE Router Static Routing or BGP Dynamic Routing MPLS MPLS 2183 MPLS Dynamic Design Model Supports up to 100 remote sites Has a single MPLS VPN carrier Uses BGP routing with MPLS VPN carrier The MPLS Dynamic design model is shown in Figure 1. Introduction December 2013 6
Dual MPLS Design Model Supports up to 500 remote sites Has multiple MPLS VPN carriers Uses BGP routing with MPLS VPN carrier Typically used with a dedicated WAN distribution layer The Dual MPLS design model is shown in the following figure. Figure 2 - Dual MPLS design model Core Layer Collapsed Core/ Distribution Layer Distribution Layer MPLS CE Routers MPLS CE Routers BGP Dynamic Routing BGP Dynamic Routing MPLS A MPLS B MPLS A MPLS B 2184 WAN Remote-Site Designs This guide documents multiple WAN remote-site designs, and they are based on various combinations of WAN transports mapped to the site specific requirements for service levels and redundancy. Figure 3 - WAN remote-site designs MPLS WAN Nonredundant Redundant Links Redundant Links & Routers MPLS MPLS-A MPLS-B MPLS-A MPLS-B 2117 Introduction December 2013 7
The remote-site designs include single or dual WAN edge routers. These are always MPLS CE routers. Most remote sites are designed with a single router WAN edge; however, certain remote-site types require a dual router WAN edge. Dual router candidate sites include regional office or remote campus locations with large user populations, or sites with business critical needs that justify additional redundancy to remove single points of failure. The overall WAN design methodology is based on a primary WAN-aggregation site design that can accommodate all of the remote-site types that map to the various link combinations listed in the following table. Table 2 - WAN remote-site transport options WAN remote-site routers WAN transports Primary transport Secondary transport Single Single MPLS VPN A Single Dual MPLS VPN A MPLS VPN B Dual Dual MPLS VPN A MPLS VPN B The modular nature of the network design enables you to create design elements that you can replicate throughout the network. Both WAN-aggregation designs and all of the WAN remote-site designs are standard building blocks in the overall design. Replication of the individual building blocks provides an easy way to scale the network and allows for a consistent deployment method. WAN/LAN Interconnection The primary role of the WAN is to interconnect primary site and remote-site LANs. The LAN discussion within this guide is limited to how the WAN-aggregation site LAN connects to the WAN-aggregation devices and how the remote-site LANs connect to the remote-site WAN devices. Specific details regarding the LAN components of the design are covered in the Campus Wired LAN Technology Design Guide. At remote sites, the LAN topology depends on the number of connected users and physical geography of the site. Large sites may require the use of a distribution layer to support multiple access-layer switches. Other sites may only require an access-layer switch directly connected to the WAN remote-site routers. The variants that are tested and documented in this guide are shown in the following table. Table 3 - WAN remote-site LAN options WAN remote-site routers WAN transports LAN topology Single Single Access only Distribution/access Single Dual Access only Distribution/access Dual Dual Access only Distribution/access Introduction December 2013 8
WAN Remote Sites LAN Topology For consistency and modularity, all WAN remote sites use the same VLAN assignment scheme, which is shown in the following table. This design guide uses a convention that is relevant to any location that has a single access switch and this model can also be easily scaled to additional access closets through the addition of a distribution layer. Table 4 - WAN remote-sites VLAN assignment VLAN Usage Layer 2 access Layer 3 distribution/ access VLAN 64 Data Yes VLAN 69 Voice Yes VLAN 99 Transit Yes (dual router only) Yes (dual router only) VLAN 50 Router link (1) Yes VLAN 54 Router link (2) Yes (dual router only) Layer 2 Access WAN remote sites that do not require additional distribution layer routing devices are considered to be flat or from a LAN perspective they are considered unrouted Layer 2 sites. All Layer 3 services are provided by the attached WAN routers. The access switches, through the use of multiple VLANs, can support services such as data and voice. The design shown in the following figure illustrates the standardized VLAN assignment scheme. The benefits of this design are clear: all of the access switches can be configured identically, regardless of the number of sites in this configuration. Access switches and their configuration are not included in this guide. The Campus Wired LAN Technology Design Guide provides configuration details on the various access switching platforms. IP subnets are assigned on a per-vlan basis. This design only allocates subnets with a 255.255.255.0 netmask for the access layer, even if less than 254 IP addresses are required. (This model can be adjusted as necessary to other IP address schemes.) The connection between the router and the access switch must be configured for 802.1Q VLAN trunking with subinterfaces on the router that map to the respective VLANs on the switch. The various router subinterfaces act as the IP default gateways for each of the IP subnet and VLAN combinations. Figure 4 - WAN remote site Flat Layer 2 LAN (single router) MPLS No HSRP Required VLAN 64 - Data VLAN 69 - Voice 802.1Q VLAN Trunk (64, 69) 2118 Introduction December 2013 9
A similar LAN design can be extended to a dual-router edge as shown in the following figure. This design change introduces some additional complexity. The first requirement is to run a routing protocol. You need to configure Enhanced Interior Gateway Protocol (EIGRP) between the routers. For consistency with the primary site LAN, use EIGRP process 100. Because there are now two routers per subnet, a First Hop Redundancy Protocol (FHRP) must be implemented. For this design, Cisco selected Hot Standby Router Protocol (HSRP) as the FHRP. HSRP is designed to allow for transparent failover of the first-hop IP router. HSRP ensures high network availability by providing first-hop routing redundancy for IP hosts configured with a default gateway IP address. HSRP is used in a group of routers for selecting an active router and a standby router. When there are multiple routers on a LAN, the active router is the router of choice for routing packets; the standby router is the router that takes over when the active router fails or when preset conditions are met. Figure 5 - WAN remote site Flat Layer 2 LAN (dual router) MPLS MPLS ibgp EIGRP VLAN99 - Transit Active HSRP Router HSRP VLANs VLAN 64 - Data VLAN 69 - Voice 802.1Q VLAN Trunk (64, 69, 99) 2119 Enhanced Object Tracking (EOT) provides a consistent methodology for various router and switching features to conditionally modify their operation based on information objects available within other processes. The objects that can be tracked include interface line protocol, IP route reachability, and IP service-level agreement (SLA) reachability, as well as several others. The IP SLA feature provides a capability for a router to generate synthetic network traffic that can be sent to a remote responder. The responder can be a generic IP endpoint that can respond to an Internet Control Message Protocol (ICMP) echo (ping) request, or can be a Cisco router running an IP SLA responder process, that can respond to more complex traffic such as jitter probes. The use of IP SLA allows the router to determine endto-end reachability to a destination and also the roundtrip delay. More complex probe types can also permit the calculation of loss and jitter along the path. IP SLA is used in tandem with EOT within this design. In order to improve convergence times after an MPLS WAN failure, HSRP has the capability to monitor the reachability of a next-hop IP neighbor through the use of EOT and IP SLA. This combination allows for a router to give up its HSRP Active role if its upstream neighbor becomes unresponsive. This provides additional network resiliency. Introduction December 2013 10
Figure 6 - WAN remote-site IP SLA probe to verify upstream device reachability Detailed View IP SLA Probe as Tracked Object WAN IP SLA Probe Upstream Interface WAN WAN WAN Interface R1 EIGRP VLAN 99 - Transit 802.1Q VLAN Trunk (64, 69, 99) Active HSRP Router HSRP VLANs VLAN 64 - Data VLAN 69 - Voice 2142 HSRP is configured to be active on the router with the highest priority WAN transport. EOT of IP SLA probes is implemented in conjunction with HSRP so that in the case of WAN transport failure, the standby HSRP router associated with the lower priority (alternate) WAN transport becomes the active HSRP router. The IP SLA probes are sent from the MPLS CE router to the MPLS Provider Edge (PE) router in order to ensure reachability of the next hop router. This is more effective than simply monitoring the status of the WAN interface. The dual router designs also warrant an additional component that is required for proper routing in certain scenarios. In these cases, a traffic flow from a remote-site host might be sent to a destination reachable via the alternate WAN transport (for example, an MPLS A + MPLS B remote site communicating with an MPLS-B-only remote site). The primary WAN transport router then forwards the traffic out the same data interface to send it to the alternate WAN transport router, which then forwards the traffic to the proper destination. This is referred to as hairpinning. The appropriate method to avoid sending the traffic out the same interface is to introduce an additional link between the routers and designate the link as a transit network (VLAN 99). There are no hosts connected to the transit network, and it is only used for router-router communication. The routing protocol runs between router subinterfaces assigned to the transit network. No additional router interfaces are required with this design modification because the 802.1Q VLAN trunk configuration can easily accommodate an additional subinterface. Distribution and Access Layer Large remote sites may require a LAN environment similar to that of a small campus LAN that includes a distribution layer and access layer. This topology works well with either a single or dual router WAN edge. To implement this design, the routers should connect via EtherChannel links to the distribution switch. These EtherChannel links are configured as 802.1Q VLAN trunks, to support both a routed point-to-point link to allow EIGRP routing with the distribution switch, and in the dual router design, to provide a transit network for direct communication between the WAN routers. Introduction December 2013 11
Figure 7 - WAN remote site Connection to distribution layer WAN WAN 802.1Q Trunk (50) 802.1Q Trunk (50, 99) 802.1Q Trunk (54, 99) 802.1Q Trunk (ww, xx) 802.1Q Trunk (yy, zz) 802.1Q Trunk (ww, xx) 802.1Q Trunk (yy, zz) VLAN 50 - Router 1 Link VLAN 50 - Router 1 Link VLAN 54 - Router 2 Link VLAN 99 - Transit 2007 The distribution switch handles all access-layer routing, with VLANs trunked to access switches. No HSRP is required when the design includes a distribution layer. A full distribution and access-layer design is shown in the following figure. Figure 8 - WAN remote site Distribution and access layer (dual router) WAN VLAN 50 - Router 1 Link VLAN 54 - Router 2 Link VLAN 99 - Transit 802.1Q Trunk (50, 99) 802.1Q Trunk (54, 99) 802.1Q Trunk (ww, xx) 802.1Q Trunk (yy, zz) VLAN ww - Data VLAN yy - Data VLAN xx - Voice No HSRP Required VLAN zz - Voice 2144 Introduction December 2013 12
IP Multicast IP Multicast allows a single IP data stream to be replicated by the infrastructure (routers and switches) and sent from a single source to multiple receivers. IP Multicast is much more efficient than multiple individual unicast streams or a broadcast stream that would propagate everywhere. IP telephony Music On Hold (MOH) and IP video broadcast streaming are two examples of IP Multicast applications. To receive a particular IP Multicast data stream, end hosts must join a multicast group by sending an Internet Group Management Protocol (IGMP) message to their local multicast router. In a traditional IP Multicast design, the local router consults another router in the network that is acting as a rendezvous point (RP) to map the receivers to active sources so that they can join their streams. The RP is a control-plane operation that should be placed in the core of the network or close to the IP Multicast sources on a pair of Layer 3 switches or routers. IP Multicast routing begins at the distribution layer if the access layer is Layer 2 and provides connectivity to the IP Multicast RP. In designs without a core layer, the distribution layer performs the RP function. This design is fully enabled for a single global scope deployment of IP Multicast. The design uses an Anycast RP implementation strategy. This strategy provides load sharing and redundancy in Protocol Independent Multicast sparse mode (PIM SM) networks. Two RPs share the load for source registration and the ability to act as hot backup routers for each other. The benefit of this strategy from the WAN perspective is that all IP routing devices within the WAN use an identical configuration referencing the Anycast RPs. IP PIM SM is enabled on all interfaces including loopbacks, VLANs, and subinterfaces. Quality of Service Most users perceive the network as just a transport utility mechanism to shift data from point A to point B as fast as it can. Many sum this up as just speeds and feeds. While it is true that IP networks forward traffic on a best-effort basis by default, this type of routing only works well for applications that adapt gracefully to variations in latency, jitter, and loss. However networks are multiservice by design and support real-time voice and video as well as data traffic. The difference is that real-time applications require packets to be delivered within specified loss, delay, and jitter parameters. In reality, the network affects all traffic flows and must be aware of end-user requirements and services being offered. Even with unlimited bandwidth, time-sensitive applications are affected by jitter, delay, and packet loss. Quality of service (QoS) enables a multitude of user services and applications to coexist on the same network. Within the architecture, there are wired and wireless connectivity options that provide advanced classification, prioritizing, queuing, and congestion mechanisms as part of the integrated QoS to help ensure optimal use of network resources. This functionality allows for the differentiation of applications, ensuring that each has the appropriate share of the network resources to protect the user experience and ensure the consistent operations of business critical applications. QoS is an essential function of the network infrastructure devices used throughout this architecture. QoS enables a multitude of user services and applications, including real-time voice, high-quality video, and delaysensitive data to coexist on the same network. In order for the network to provide predictable, measurable, and sometimes guaranteed services, it must manage bandwidth, delay, jitter, and loss parameters. Even if you do not require QoS for your current applications, you can use QoS for management and network protocols to protect the network functionality and manageability under normal and congested traffic conditions. The goal of this design is to provide sufficient classes of service to allow you to add voice, interactive video, critical data applications, and management traffic to the network, either during the initial deployment or later with minimum system impact and engineering effort. Introduction December 2013 13
The QoS classifications in the following table are applied throughout this design. This table is included as a reference. Table 5 - QoS service class mappings Service class Per-hop behavior (PHB) Differentiated services code point (DSCP) IP precedence (IPP) Network layer Layer 3 Layer 3 Layer 3 Layer 2 Network control CS6 48 6 6 Telephony EF 46 5 5 Signaling CS3 24 3 3 Multimedia conferencing AF41, 42, 43 34, 36, 38 4 4 Real-time interactive CS4 32 4 4 Multimedia streaming AF31, 32, 33 26, 28, 30 3 3 Broadcast video CS5 40 4 4 Low-latency data AF21, 22, 23 18, 20, 22 2 2 Operation, administration, and maintenance (OAM) CS2 16 2 2 Bulk data AF11, 12, 13 10, 12, 14 1 1 Scavenger CS1 8 1 1 Default best effort DF 0 0 0 Class of service (CoS) Introduction December 2013 14
Deploying the WAN Overall WAN Architecture Design Goals IP Routing The design has the following IP routing goals: Provide optimal routing connectivity from primary WAN-aggregation sites to all remote locations Isolate WAN routing topology changes from other portions of the network Ensure active/standby symmetric routing when multiple paths exist, for ease of troubleshooting and to prevent oversubscription of IP telephony Call Admission Control (CAC) limits Provide site-site remote routing via the primary WAN-aggregation site (hub-and-spoke model) Permit optimal direct site-site remote routing when carrier services allow (spoke-to-spoke model) Support IP Multicast sourced from the primary WAN-aggregation site At the WAN remote sites, there is no local Internet access for web browsing or cloud services. This model is referred to as a centralized Internet model. It is worth noting that sites with Internet/DMVPN for backup transport could potentially provide local Internet capability; however, for this design, only encrypted traffic to other DMVPN sites is permitted to use the Internet link. In the centralized Internet model, a default route is advertised to the WAN remote sites in addition to the internal routes from the data center and campus. LAN Access All remote sites are to support both wired LAN access. High Availability The network must tolerate single failure conditions including the failure of any single WAN transport link or any single network device at the primary WAN-aggregation site. Remote sites classified as single-router, dual-link must be able tolerate the loss of either WAN transport. Remote sites classified as dual-router, dual-link must be able to tolerate the loss of either an edge router or a WAN transport. Path Selection Preferences There are many potential traffic flows based on which WAN transports are in use and whether or not a remote site is using a dual WAN transport. The single WAN transport routing functions as follows. MPLS VPN-connected site: Connects to a site on the same MPLS VPN The optimal route is direct within the MPLS VPN (traffic is not sent to the primary site). Connects to any other site The route is through the primary site. The use of the dual WAN transports is specifically tuned to behave in an active/standby manner. This type of configuration provides symmetric routing, with traffic flowing along the same path in both directions. Symmetric routing simplifies troubleshooting because bidirectional traffic flows always traverse the same links. Deploying the WAN December 2013 15
The design assumes that one of the MPLS VPN WAN transports is designated as the primary transport, which is the preferred path in most conditions. MPLS VPN primary + MPLS VPN secondary dual-connected site: Connects to a site on the same MPLS VPN The optimal route is direct within the MPLS VPN (traffic is not sent to the primary site). Connects to any other site The route is through the primary site. Quality of Service (QoS) The network must ensure that business applications perform across the WAN during times of network congestion. Traffic must be classified and queued and the WAN connection must be shaped to operate within the capabilities of the connection. When the WAN design uses a service provider offering with QoS, the WAN edge QoS classification and treatment must align to the service provider offering to ensure consistent end-toend QoS treatment of traffic. Design Parameters This design guide uses certain standard design parameters and references various network infrastructure services that are not located within the WAN. These parameters are listed in the following table. Table 6 - Universal design parameters Network service IP address Domain name cisco.local Active Directory, DNS server, DHCP server 10.4.48.10 Cisco Secure Access Control System (ACS) 10.4.48.15 Network Time Protocol (NTP) server 10.4.48.17 Deploying the WAN December 2013 16
Deploying an MPLS WAN Design Overview WAN-Aggregation MPLS CE Routers The MPLS WAN designs are intended to support up to 500 remote sites with a combined aggregate WAN bandwidth of up to 1.0 Gbps. The most critical devices are the WAN routers that are responsible for reliable IP forwarding and QoS. The amount of bandwidth required at the WAN-aggregation site determines which model of router to use. The choice of whether to implement a single router or dual router is determined by the number of carriers that are required in order to provide connections to all of the remote sites. Cisco ASR 1000 Series Aggregation Services Routers represent the next-generation, modular, servicesintegrated Cisco routing platform. They are specifically designed for WAN aggregation, with the flexibility to support a wide range of 3- to 16-mpps (millions of packets per second) packet-forwarding capabilities, 2.5- to 40-Gbps system bandwidth performance, and scaling. The Cisco ASR 1000 Series is fully modular from both hardware and software perspectives, and the routers have all the elements of a true carrier-class routing product that serves both enterprise and service-provider networks. This design uses the following routers as MPLS CE routers: Cisco ASR 1002-X router configured with an embedded service processor (ESP) default bandwidth of 5 Gbps upgradable with software licensing options to 10 Gbps, 20 Gbps, and 36 Gbps Cisco ASR 1002 router configured with an embedded service processor 5 (ESP5) Cisco ASR 1001 router fixed configuration with a 2.5 Gbps embedded service processor Cisco 3945 Integrated Services Router Cisco 3925 Integrated Services Router All of the design models can be constructed using any of the MPLS CE routers listed in Table 7. You should consider the following: the forwarding performance of the router using an Ethernet WAN deployment with broad services enabled, the router s alignment with the suggested design model, and the number of remote sites. Table 7 - WAN aggregation MPLS CE router options Service Cisco 3925 Cisco 3945 ASR 1001 ASR 1002 ASR 1002-X Ethernet WAN with services 100 Mbps 150 Mbps 250 Mbps 500 Mbps 500Mbps-1.5Gbps Software Redundancy Option None None Yes Yes Yes Redundant power supply Option Option Default Default Default Supported Design Models All All All All All Suggested Design Model MPLS Static MPLS Static MPLS Dynamic Dual MPLS Dual MPLS Suggested Number of Remote Sites 25 50 100 250 250+ 17
Remote Sites MPLS CE Router Selection The actual WAN remote-site routing platforms remain unspecified because the specification is tied closely to the bandwidth required for a location and the potential requirement for the use of service module slots. The ability to implement this solution with a variety of potential router choices is one of the benefits of a modular design approach. There are many factors to consider in the selection of the WAN remote-site routers. Among those, and key to the initial deployment, is the ability to process the expected amount and type of traffic. You also need to make sure that you have enough interfaces, enough module slots, and a properly licensed Cisco IOS Software image that supports the set of features that is required by the topology. Cisco tested multiple integrated service router models as MPLS CE routers, and the expected performance is shown in the following table. Table 8 - WAN remote-site Cisco Integrated Services Router options 881V 1 1941 2 2911 2921 2951 3925 3945 4451-X Ethernet WAN with services 3 8 Mbps 25 Mbps 35 Mbps 50 Mbps 75 Mbps 100 Mbps 150 Mbps 1 Gbps On-board FE ports 1 (and 4-port LAN switch) 0 0 0 0 0 0 0 On-board GE ports 4 0 2 3 3 3 3 3 4 Service module slots 5 0 0 1 1 2 2 4 2 Redundant power supply option No No No No No Yes Yes Yes Notes: 1. The Cisco 881 Integrated Services Router is recommended for use at single-router, single-link remote sites with VoIP requirements. 2. The 1941 is recommended for use at single-router, single-link remote sites. 3. The performance numbers are conservative numbers obtained when the router is passing IMIX traffic with heavy services configured and the CPU utilization is under 75 percent. 4. A single-router, dual-link remote-site requires four router interfaces when using a port-channel to connect to an access or distribution layer. Add the EHWIC-1GE-SFP-CU to the Cisco 2900 and 3900 Series Integrated Services Routers in order to provide the additional WAN-facing interface. 5. Not all service modules are supported in Cisco 4451-X ISR. Some service modules are double-wide. The MPLS CE routers at the WAN remote sites connect in the same manner as the MPLS CE routers at the WAN-aggregation site. The single link MPLS WAN remote site is the most basic of building blocks for any remote location. You can use this design with the CE router connected directly to the access layer, or you can use it to support a more complex LAN topology by connecting the CE router directly to a distribution layer. The IP routing is straightforward and can be handled entirely by using static routes at the WAN-aggregation site and static default routes at the remote site. However, there is significant value to configuring this type of site with dynamic routing and this approach is used for the MPLS Dynamic and Dual MPLS designs. Dynamic routing makes it easy to add or modify IP networks at the remote site because any changes are immediately propagated to the rest of the network. MPLS VPN-connected sites require static routing in order to be handled by the carrier, and any changes or modifications require a change request to the carrier. The smaller scale MPLS Static design uses static routing and relies on the carrier to configure the additional required static routes on the PE routers. 18
Tech Tip We recommend that you select the Dual MPLS or MPLS Dynamic designs if you intend to use resilient WAN links or want to be able to modify your routing configuration without carrier involvement. Figure 9 - MPLS WAN remote site (single-router, single-link) MPLS VPN MPLS VPN Static Routing Dynamic Routing Static Routing MPLS Dynamic MPLS Static 2124 You can augment the basic single-link design by adding an alternate WAN transport that uses a secondary MPLS carrier and either connects on the same router or on an additional router. By adding an additional link, you provide the first level of high availability for the remote site. The router can automatically detect failure of the primary link and reroute traffic to the secondary path. It is mandatory to run dynamic routing when there are multiple paths and the Dual MPLS or MPLS Dynamic design models are used. The routing protocols are tuned to ensure the proper path selection. Figure 10 - MPLS WAN dual-carrier remote site (dual-link options) MPLS VPN A MPLS VPN B MPLS VPN A MPLS VPN B Dual MPLS Design Model Only 2125 The dual-router, dual-link design continues to improve upon the level of high availability for the site. This design can tolerate the loss of the primary router because the secondary router reroutes traffic via the alternate path. Design Details All WAN-aggregation MPLS CE routers connect to the same resilient switching device in the distribution layer. All devices use EtherChannel connections consisting of two port bundles. This design provides both resiliency and additional forwarding performance. You can accomplish additional forwarding performance by increasing the number of physical links within an EtherChannel. 19
WAN transport via Ethernet is the only media type tested and included in the configuration section. Other media types are commonly used (such as T1/E1), and these technologies are reliable and well understood. Due to the multiplicity of potential choices for transport, media type, and interface type, we decided to limit the focus of this design guide. Documentation of additional variants is available in other guides. MPLS VPNs require a link between a PE router and a CE router. The PE and CE routers are considered IP neighbors across this link. CE routers are only able to communicate with other CE routers across the WAN via intermediate PE routers. Figure 11 - MPLS VPN (PE-CE connections) MPLS Carrier CE PE PE CE Direct Adjacencies Only Between CE and PE Routers Direct Adjacencies Only Between CE and PE Routers 2126 Both the PE and CE routers are required to have sufficient IP-routing information in order to provide end-toend reachability. To maintain this routing information, you typically need to use a routing protocol; BGP is most commonly used for this purpose. The various CE routers advertise their routes to the PE routers. The PE routers propagate the routing information within the carrier network and in turn re-advertise the routes back to other CE routers. This propagation of routing information is known as dynamic PE-CE routing and it is essential when any sites have multiple WAN transports (often referred to as dual-homed or multi-homed). The Dual MPLS and MPLS Dynamic designs use dynamic PE-CE routing with BGP. Tech Tip EIGRP and Open Shortest Path First (OSPF) Protocol are also effective as PE-CE routing protocols, but may not be universally available across all MPLS VPN carriers. Sites with only a single WAN transport (a single-homed site) do not require dynamic PE-CE routing, and can rely on static routing because there is only a single path to any destination. This design recommends dynamic PE-CE routing to provide consistency with configurations across both single-homed and dual-homed sites. This also allows for easy transition from a single-homed to a dual-homed remote-site design by adding an additional link to an existing remote site. A static routing option is also included to support smaller scale requirements that do not require a dynamic routing protocol. Static routing is used in the MPLS Static design model. Cisco did not test the PE routers, and their configurations are not included in this guide. For an MPLS VPN WAN deployment, you need to install and configure MPLS CE routers at every location, including the WAN-aggregation site, and at every MPLS WAN-connected remote site. At the WAN-aggregation site, an MPLS CE router must be connected both to the distribution layer and to its respective MPLS carrier. Multiple routing protocols (EIGRP and BGP) are used to exchange routing information, and the routing protocol configurations are tuned from their default settings to influence traffic flows to their desired behavior. The IP routing details for the single and dual MPLS carrier WAN-aggregation topology with dynamic routing are shown in the following figure. 20
Figure 12 - Dual MPLS and MPLS Dynamic designs MPLS CE routing detail WAN Distribution WAN Distribution EIGRP EIGRP EIGRP MPLS CE Routers ibgp MPLS CE Routers ebgp ebgp ebgp MPLS A MPLS B MPLS Dual MPLS MPLS Dynamic 2127 The IP routing details for the single MPLS carrier WAN-aggregation topology with static routing are shown in the following figure. Figure 13 - MPLS Static Design MPLS CE routing detail WAN Distribution EIGRP MPLS CE Router Static Routing MPLS Static Routing 2128 EIGRP Cisco chose EIGRP as the primary routing protocol because it is easy to configure, does not require a large amount of planning, has flexible summarization and filtering, and can scale to large networks. As networks grow, the number of IP prefixes or routes in the routing tables grows as well. You should program IP summarization on links where logical boundaries exist, such as distribution layer links to the wide area or to a core. By performing IP summarization, you can reduce the amount of bandwidth, processor, and memory necessary to carry large route tables, and reduce convergence time associated with a link failure. In this design, EIGRP process 100 is the primary EIGRP process and is referred to as EIGRP-100. EIGRP-100 is used at the WAN-aggregation site to connect to the primary site LAN distribution layer and at WAN remote sites with dual WAN routers or with distribution-layer LAN topologies. 21
BGP Cisco chose BGP as the routing protocol for PE and CE routers to connect to the MPLS VPNs because it is consistently supported across virtually all MPLS carriers. In this role, BGP is straightforward to configure and requires little or no maintenance. BGP scales well and you can use it to advertise IP aggregate addresses for remote sites. To use BGP, you must select an Autonomous System Number (ASN). In this design, we use a private ASN (65511) as designated by the Internet Assigned Numbers Authority (IANA). The private ASN range is 64512 to 65534. A dual-carrier MPLS design requires an ibgp connection between the CE routers to properly retain routing information for the remote sites. Deployment Details The procedures in this section provide examples for some settings. The actual settings and values that you use are determined by your current network configuration. Table 9 - Parameters used in the deployment examples Hostname Loopback IP Address Port Channel IP Address CE-ASR1002-1 10.4.32.241/32 10.4.32.2/30 CE-ASR1001-2 10.4.32.242/32 10.4.32.6/30 Configuring the MPLS CE Router PROCESS 1. Configure the distribution switch 2. Configure the WAN Aggregation Platform 3. Configure connectivity to the LAN 4. Connect to MPLS PE router 5. Redistribute WAN routes into EIGRP 6. Configure BGP Procedure 1 Configure the distribution switch Reader Tip This process assumes that the distribution switch has already been configured following the guidance in the Campus Wired LAN Technology Design Guide. Only the procedures required to support the integration of the WAN-aggregation router into the deployment are included. 22
The LAN distribution switch is the path to the organization s main campus and data center. A Layer 3 portchannel interface connects to the distribution switch to the WAN-aggregation router and the internal routing protocol peers across this interface. Tech Tip As a best practice, use the same channel numbering on both sides of the link where possible. Step 1: Configure the Layer 3 port-channel interface and assign the IP address. interface Port-channel1 description CE-ASR1002-1 no switchport ip address 10.4.32.1 255.255.255.252 ip pim sparse-mode logging event link-status carrier-delay msec 0 no shutdown Step 2: Configure EtherChannel member interfaces. Configure the physical interfaces to tie to the logical port-channel by using the channel-group command. The number for the port-channel and channel-group must match. Not all router platforms can support Link Aggregation Control Protocol (LACP) to negotiate with the switch, so you configure EtherChannel statically. Also, apply the egress QoS macro that was defined in the platform configuration procedure to ensure traffic is prioritized appropriately. interface GigabitEthernet1/0/1 description CE-ASR1002-1 Gig0/0/0 interface GigabitEthernet2/0/1 description CE-ASR1002-1 Gig0/0/1 interface range GigabitEthernet1/0/1, GigabitEthernet2/0/1 no switchport macro apply EgressQoS carrier-delay msec 0 channel-group 1 mode on logging event link-status logging event trunk-status logging event bundle-status no shutdown Step 3: Allow the routing protocol to form neighbor relationships across the port channel interface. router eigrp 100 no passive-interface Port-channel1 23
Step 4: On the distribution layer switch, configure the layer 3 interface connected to the LAN core to summarize the WAN network range. Tech Tip It is a best practice to summarize IP routes from the WAN distribution layer towards the core. interface Port-channel38 description Link to C6500-VSS ip summary-address eigrp 100 10.4.32.0 255.255.248.0 ip summary-address eigrp 100 10.4.128.0 255.255.240.0 ip summary-address eigrp 100 10.4.160.0 255.255.252.0 ip summary-address eigrp 100 10.5.0.0 255.255.0.0 Step 5: On the distribution layer switch, configure the layer 3 interfaces connected to the WAN aggregation routers to summarize the WAN remote-site network range. Tech Tip It is a best practice to summarize IP routes from the WAN distribution layer towards the MPLS WAN. interface Port-channel1 description CE-ASR1002-1 ip summary-address eigrp 100 10.5.0.0 255.255.0.0 Repeat this step as needed for additional WAN aggregation routers. Procedure 2 Configure the WAN Aggregation Platform Within this design, there are features and services that are common across all WAN aggregation routers. These are system settings that simplify and secure the management of the solution. Step 1: Configure the device host name. This makes it easy to identify the device. hostname CE-ASR1002-1 24
Step 2: Configure local login and password. The local login account and password provides basic access authentication to a router, which provides only limited operational privileges. The enable password secures access to the device configuration mode. By enabling password encryption, you prevent the disclosure of plain text passwords when viewing configuration files. username admin password c1sco123 enable secret c1sco123 service password-encryption aaa new-model Step 3: By default, HTTPS access to the router uses the enable password for authentication. Step 4: (Optional) Configure centralized user authentication. As networks scale in the number of devices to maintain it poses an operational burden to maintain local user accounts on every device. A centralized authentication, authorization, and accounting (AAA) service reduces operational tasks per device and provides an audit log of user access for security compliance and root cause analysis. When AAA is enabled for access control, all management access to the network infrastructure devices (SSH and HTTPS) is controlled by AAA. TACACS+ is the primary protocol used to authenticate management logins on the infrastructure devices to the AAA server. A local AAA user database is also defined in Step 2 on each network infrastructure device to provide a fallback authentication source in case the centralized TACACS+ server is unavailable. tacacs server TACACS-SERVER-1 address ipv4 10.4.48.15 key SecretKey aaa group server tacacs+ TACACS-SERVERS server name TACACS-SERVER-1 aaa authentication login default group TACACS-SERVERS local aaa authorization exec default group TACACS-SERVERS local aaa authorization console ip http authentication aaa Step 5: Configure device management protocols. Secure HTTP (HTTPS) and Secure Shell (SSH) are secure replacements for the HTTP and Telnet protocols. They use Secure Sockets Layer (SSL) and Transport Layer Security (TLS) to provide device authentication and data encryption. Secure management of the network device is enabled through the use of the SSH and HTTPS protocols. Both protocols are encrypted for privacy and the unsecure protocols, Telnet and HTTP, are turned off. 25
Specify the transport preferred none on vty lines to prevent errant connection attempts from the CLI prompt. Without this command, if the ip name-server is unreachable, long timeout delays may occur for mistyped commands. ip domain-name cisco.local ip ssh version 2 no ip http server ip http secure-server line vty 0 15 transport input ssh transport preferred none Step 6: Enable synchronous logging. When synchronous logging of unsolicited messages and debug output is turned on, console log messages are displayed on the console after interactive CLI output is displayed or printed. With this command, you can continue typing at the device console when debugging is enabled. line con 0 logging synchronous Step 7: Enable Simple Network Management Protocol (SNMP). This allows the network infrastructure devices to be managed by a Network Management System (NMS). SNMPv2c is configured both for a read-only and a read-write community string. snmp-server community cisco RO snmp-server community cisco123 RW Step 8: If operational support is centralized in your network, you can increase network security by using an access list to limit the networks that can access your device. In this example, only devices on the 10.4.48.0/24 network will be able to access the device via SSH or SNMP. access-list 55 permit 10.4.48.0 0.0.0.255 line vty 0 15 access-class 55 in snmp-server community cisco RO 55 snmp-server community cisco123 RW 55 Tech Tip If you configure an access-list on the vty interface you may lose the ability to use ssh to login from one router to the next for hop-by-hop troubleshooting. Step 9: Configure a synchronized clock. The Network Time Protocol (NTP) is designed to synchronize a network of devices. An NTP network usually gets its time from an authoritative time source, such as a radio clock or an atomic clock attached to a time server. NTP then distributes this time across the organizations network. 26
You should program network devices to synchronize to a local NTP server in the network. The local NTP server typically references a more accurate clock feed from an outside source. By configuring console messages, logs, and debug output to provide time stamps on output, you can cross-reference events in a network. ntp server 10.4.48.17 clock timezone PST -8 clock summer-time PDT recurring service timestamps debug datetime msec localtime service timestamps log datetime msec localtime Step 10: Configure an in-band management interface. The loopback interface is a logical interface that is always reachable as long as the device is powered on and any IP interface is reachable to the network. Because of this capability, the loopback address is the best way to manage the switch in-band. Layer 3 process and features are also bound to the loopback interface to ensure process resiliency. The loopback address is commonly a host address with a 32-bit address mask. Allocate the loopback address from the IP address block that the distribution switch summarizes to the rest of the network. interface Loopback 0 ip address 10.4.32.241 255.255.255.255 ip pim sparse-mode Step 11: Bind the device processes for SNMP, SSH, PIM, TACACS+ and NTP to the loopback interface address. This provides optimal resiliency: snmp-server trap-source Loopback0 ip ssh source-interface Loopback0 ip pim register-source Loopback0 ip tacacs source-interface Loopback0 ntp source Loopback0 Step 12: Configure IP unicast routing. EIGRP is configured facing the LAN distribution or core layer. In this design, the port-channel interface and the loopback must be EIGRP interfaces. The loopback may remain a passive interface. The network range must include both interface IP addresses, either in a single network statement or in multiple network statements. This design uses a best practice of assigning the router ID to a loopback address. router eigrp 100 network 10.4.0.0 0.1.255.255 no auto-summary passive-interface default eigrp router-id 10.4.32.241 Step 13: Configure IP Multicast routing. IP Multicast allows a single IP data stream to be replicated by the infrastructure (routers and switches) and sent from a single source to multiple receivers. Using IP Multicast is much more efficient than using multiple individual unicast streams or a broadcast stream that would propagate everywhere. IP Telephony MOH and IP Video Broadcast Streaming are two examples of IP Multicast applications. 27
In order to receive a particular IP Multicast data stream, end hosts must join a multicast group by sending an IGMP message to their local multicast router. In a traditional IP Multicast design, the local router consults another router in the network that is acting as an RP to map the receivers to active sources so they can join their streams. This design, which is based on sparse mode multicast operation, uses Auto RP for a simple yet scalable way to provide a highly resilient RP environment. Enable IP Multicast routing on the platforms in the global configuration mode. ip multicast-routing If you are using a Cisco ASR 1000 Series router, the distributed keyword is required. ip multicast-routing distributed Step 14: Configure every Layer 3 switch and router to discover the IP Multicast RP with autorp. Use the ip pim autorp listener command to allow for discovery across sparse mode links. This configuration provides for future scaling and control of the IP Multicast environment and can change based on network needs and design. ip pim autorp listener Step 15: Enable sparse mode multicast operation for all Layer 3 interfaces in the network. ip pim sparse-mode Procedure 3 Configure connectivity to the LAN Any links to adjacent distribution layers should be Layer 3 links or Layer 3 EtherChannels. Step 1: Configure Layer 3 interface. interface Port-channel1 ip address 10.4.32.2 255.255.255.252 ip pim sparse-mode no shutdown Step 2: Configure EtherChannel member interfaces. Configure the physical interfaces to tie to the logical port-channel by using the channel-group command. The number for the port-channel and channel-group must match. Not all router platforms can support LACP to negotiate with the switch, so you configure EtherChannel statically. interface GigabitEthernet0/0/0 description WAN-D3750X Gig1/0/1 interface GigabitEthernet0/0/1 description WAN-D3750X Gig2/0/1 interface range GigabitEthernet0/0/0, GigabitEthernet0/0/1 no ip address channel-group 1 no shutdown 28
Step 3: Configure the EIGRP interface. Allow EIGRP to form neighbor relationships across the interface to establish peering adjacencies and exchange route tables. router eigrp 100 no passive-interface Port-channel1 Procedure 4 Connect to MPLS PE router Step 1: Assign the interface bandwidth. The bandwidth value should correspond to the actual interface speed. Or, if you are using a subrate service, use the policed rate from the carrier. The example shows a Gigabit interface (1000 Mbps) with a subrate of 300 Mbps. interface GigabitEthernet0/0/3 bandwidth 300000 Tech Tip Command reference: bandwidth kbps (300 Mbps = 300,000 kbps) Step 2: Assign the IP address and netmask of the WAN interface. The IP addressing used between CE and PE routers must be negotiated with your MPLS carrier. Typically a pointto-point netmask of 255.255.255.252 is used. interface GigabitEthernet0/0/3 ip address 192.168.3.1 255.255.255.252 Step 3: Administratively enable the interface and disable CDP. We do not recommend the use of CDP on external interfaces. interface GigabitEthernet0/0/3 no cdp enable no shutdown Procedure 5 Redistribute WAN routes into EIGRP The WAN-aggregation CE routers are configured either for dynamic routing with BGP or are statically routed. If you have a remote-site design that includes sites with dual WAN links, or do not wish to have your MPLS carrier make changes or modifications, then use the BGP option. This is the recommended approach. If your remote-site design only uses single WAN links and you don t anticipate adding or modifying IP networks at the remote sites, then you can use the statically routed option. The MPLS carrier is responsible for configuring static IP routing within the MPLS network. 29
Tech Tip If you do not use dynamic routing with BGP, then the MPLS carrier must configure a set of static routes on its PE routers for the WAN-aggregation site and for each of the remote sites. Site-specific routing details must be shared with your MPLS carrier. Option 1: BGP dynamic routing with MPLS carrier Step 1: Redistribute BGP into EIGRP. A default metric redistributes the BGP routes into EIGRP. By default, only the bandwidth and delay values are used for metric calculation. router eigrp [as number] default-metric [bandwidth (Kbps)] [delay (usec)] 255 1 1500 redistribute bgp [BGP ASN] Step 2: Configure route-map and inbound distribute-list for EIGRP. This design uses mutual route redistribution: BGP routes are distributed into EIGRP and EIGRP routes are distributed into BGP (covered inprocedure 6). It is important to tightly control how routing information is shared between different routing protocols when you use this configuration; otherwise, you might experience route flapping, where certain routes are repeatedly installed and withdrawn from the device routing tables. Proper route control ensures the stability of the routing table. An inbound distribute-list with a route-map is used to limit which routes are accepted for installation into the route table. The WAN-aggregation MPLS CE routers are configured to only accept routes that do not originate from the MPLS or DMVPN WAN sources. To accomplish this task, you must create a route-map that matches any routes originating from the WAN indicated by a specific route tag. This method allows for dynamic identification of the various WAN routes. BGP-learned routes are implicitly tagged with their respective source AS and other WAN routes are explicitly tagged by their WAN-aggregation router (documented in a separate procedure). The specific route tags in use are shown below. Table 10 - Route tag information for WAN-aggregation MPLS CE routers Tag Route source Tag method action 65401 MPLS VPN A implicit block 65402 MPLS VPN B implicit block 300 Layer 2 WAN explicit accept 65512 DMVPN hub routers explicit block This example includes all WAN route sources in the reference design. Depending on the actual design of your network, you may need to block more tags. It is important when creating the route-map that you include a permit statement at the end in order to permit the installation of routes with non-matching tags. 30
Tech Tip If you configure mutual route redistribution without proper matching, tagging, and filtering, route-flapping may occur, which can cause instability. route-map BLOCK-TAGGED-ROUTES deny 10 match tag 65401 65402 65512 route-map BLOCK-TAGGED-ROUTES permit 20 router eigrp 100 distribute-list route-map BLOCK-TAGGED-ROUTES in default-metric 100000 100 255 1 1500 redistribute bgp 65511 Option 2: Static routing with service provider Step 1: Configure static routes to remote sites LANs on the WAN-aggregation CE router. It is a best practice to summarize the remote-site network ranges into a single route when possible. ip route 10.5.0.0 255.255.0.0 192.168.3.2 Step 2: It is desirable to advertise a route for the MPLS PE-CE links, which includes the CE routers WAN interfaces, so you can use this to determine router reachability, for troubleshooting. It is a best practice to summarize the PE-CE link ranges into a single route when possible. ip route 192.168.3.0 255.255.255.0 192.168.3.2 Step 3: Configure routes to the remote-site router loopback addresses. A single summary route for the loopback range may be used when possible. ip route 10.255.251.0 255.255.255.0 192.168.3.2 Step 4: Configure EIGRP to advertise the remote-site static routes. A default metric redistributes these routes into EIGRP. By default, only the bandwidth and delay values are used for metric calculation. router eigrp [as number] default-metric [bandwidth (Kbps)] [delay (usec)] 255 1 1500 redistribute static 31
Procedure 6 Configure BGP If you are using BGP dynamic routing with the MPLS carrier, complete this procedure. Step 1: Enable BGP. To complete this step, you must use a BGP ASN. You can consult with your MPLS carrier on the requirements for the ASN, but you may be permitted to use a private ASN as designated by IANA. The private ASN range is 64512 to 65534. router bgp 65511 no synchronization bgp router-id 10.4.32.241 bgp log-neighbor-changes no auto-summary Step 2: Configure ebgp. You must configure BGP with the MPLS carrier PE device. The MPLS carrier must provide their ASN (the ASN in the previous step is the ASN identifying your site). Because the carrier PE router uses a different ASN, this configuration is considered an external BGP (ebgp) connection. The CE router advertises only network routes to the PE via BGP when: The route is specified in network statements and is present in the local routing table. The route is redistributed into BGP. It is desirable to advertise a route for the PE-CE link, so you should include this network in a network statement. You can use this to determine router reachability, for troubleshooting. router bgp 65511 network 192.168.3.0 mask 255.255.255.252 neighbor 192.168.3.2 remote-as 65401 Step 3: Redistribute EIGRP into BGP. All EIGRP routes learned by the CE router, including routes from the core and for other WAN sites, should be advertised into the WAN. It is most efficient if you summarize these routes before they are advertised to the CE router. Because BGP does not propagate a default route via redistribution, you must explicitly specify 0.0.0.0 in a network statement. router bgp 65511 network 0.0.0.0 redistribute eigrp 100 Step 4: If you have dual MPLS carriers, configure a BGP link between the CE routers. Because the CE routers are using the same ASN, this configuration is considered an internal BGP (ibgp) connection. This design uses ibgp peering using device loopback addresses, which requires the update-source and next-hop-self-configuration options. router bgp 65511 neighbor 10.4.32.242 remote-as 65511 neighbor 10.4.32.242 update-source Loopback0 neighbor 10.4.32.242 next-hop-self 32
Configuring the Remote-Site MPLS CE Router 1. Configure the WAN Remote Router 2. Connect to the MPLS PE Router PROCESS 3. Configure WAN routing 4. Connect router to access-layer switch 5. Configure access-layer routing 6. Configure remote-site DHCP 7. Configure access-layer HSRP 8. Configure the transit network 9. Configure EIGRP (LAN side) 10. Configure BGP 11. Enable Enhanced Object Tracking Use this process for the configuration of any of the following: MPLS CE router for an MPLS WAN remote site (single router, single link) MPLS WAN Dual Carrier remote site Use the following procedures when performing the initial configuration of a dual-connected MPLS CE in the single-router, dual-link design or for configuring the first router of the dual-router, dual-link design. 33
The following flowchart provides details about the configuration process for a remote-site MPLS CE router. Figure 14 - Remote-site MPLS CE router configuration flowchart Remote-Site MPLS CE Router Single Router, Single Link Remote-Site MPLS CE Router Single Router, Dual Link Remote-Site MPLS CE Router Dual Router, Dual Link (1st Router) Remote-Site MPLS CE Router Configuration Procedures 1. Configure the WAN Remote Router 2. Connect to MPLS PE Router 3. Configure WAN Routing NO Distribution Layer Design? YES Remote-Site Router to Distribution Layer Procedures 4. Connect Router to Access Layer Switch 1. Connect to Distribution Layer 2. Configure EIGRP (LAN Side) NO Dual Router Design? 5. Configure Access Layer Routing 6. Configure Remote-Site DHCP (Optional) YES NO Dual Router Design? YES 7. Configure Access Layer HSRP 8. Configure Transit Network 9. Configure EIGRP (LAN Side) ) 10. Configure BGP for Dual Router 11. Enable Enhanced Object O Tracking 3. ConfigureTransitNetwork 4. Configure BGP for Dual Router MPLS CE Router Configuration Complete MPLS CE Spoke Router Configuration Complete Configure Second Remote-Site Router Configure Second Remote-Site Router 2129 34
Procedure 1 Configure the WAN Remote Router Within this design, there are features and services that are common across all WAN remote-site routers. These are system settings that simplify and secure the management of the solution. Step 1: Configure the device host name to make it easy to identify the device. hostname [hostname] Step 2: Configure the local login and password. The local login account and password provides basic access authentication to a router that provides only limited operational privileges. The enable password secures access to the device configuration mode. By enabling password encryption, you prevent the disclosure of plain text passwords when viewing configuration files. username admin password c1sco123 enable secret c1sco123 service password-encryption aaa new-model Step 3: By default, https access to the router uses the enable password for authentication. Step 4: (Optional) Configure centralized user authentication. As networks scale in the number of devices to maintain, it can be an operational burden to maintain local user accounts on every device. A centralized authentication, authorization, and accounting (AAA) service reduces operational tasks per device and provides an audit log of user access for security compliance and root cause analysis. When AAA is enabled for access control, all management access to the network infrastructure devices (SSH and HTTPS) is controlled by AAA. TACACS+ is the primary protocol used to authenticate management logins on the infrastructure devices to the AAA server. A local AAA user database is also defined in Step 2 on each network infrastructure device in order to provide a fallback authentication source in case the centralized TACACS+ server is unavailable. tacacs server TACACS-SERVER-1 address ipv4 10.4.48.15 key SecretKey aaa group server tacacs+ TACACS-SERVERS server name TACACS-SERVER-1 aaa authentication login default group TACACS-SERVERS local aaa authorization exec default group TACACS-SERVERS local aaa authorization console ip http authentication aaa Step 5: Configure device management protocols. Secure HTTP (HTTPS) and Secure Shell (SSH) are secure replacements for the HTTP and Telnet protocols. They use Secure Sockets Layer (SSL) and Transport Layer Security (TLS) to provide device authentication and data encryption. Secure management of the network device is enabled through the use of the SSH and HTTPS protocols. Both protocols are encrypted for privacy and the unsecure protocols, Telnet and HTTP, are turned off. 35
Specify the transport preferred none on vty lines to prevent errant connection attempts from the CLI prompt. Without this command, if the ip name-server is unreachable, long timeout delays may occur for mistyped commands. ip domain-name cisco.local ip ssh version 2 no ip http server ip http secure-server line vty 0 15 transport input ssh transport preferred none Step 6: Enable synchronous logging. When synchronous logging of unsolicited messages and debug output is turned on, console log messages are displayed on the console after interactive CLI output is displayed or printed. With this command, you can continue typing at the device console when debugging is enabled. line con 0 logging synchronous Step 7: Enable Simple Network Management Protocol (SNMP). This allows the network infrastructure devices to be managed by a Network Management System (NMS). Configure SNMPv2c both for a read-only and a readwrite community string. snmp-server community cisco RO snmp-server community cisco123 RW Step 8: If your network operational support is centralized, you can increase network security by using an access list to limit the networks that can access your device. In this example, only devices on the 10.4.48.0/24 network will be able to access the device via SSH or SNMP. access-list 55 permit 10.4.48.0 0.0.0.255 line vty 0 15 access-class 55 in snmp-server community cisco RO 55 snmp-server community cisco123 RW 55 Tech Tip If you configure an access-list on the vty interface you may lose the ability to use ssh to log in from one router to the next for hop-by-hop troubleshooting. Step 9: Configure a synchronized clock. The Network Time Protocol (NTP) is designed to synchronize a network of devices. An NTP network usually gets its time from an authoritative time source, such as a radio clock or an atomic clock attached to a time server. NTP then distributes this time across the organization s network. 36
You should program network devices to synchronize to a local NTP server in the network. The local NTP server typically references a more accurate clock feed from an outside source. By configuring console messages, logs, and debug output to provide time stamps on output, you can cross-reference events in a network. ntp server 10.4.48.17 ntp update-calendar clock timezone PST -8 clock summer-time PDT recurring service timestamps debug datetime msec localtime service timestamps log datetime msec localtime Step 10: Configure an in-band management interface. The loopback interface is a logical interface that is always reachable as long as the device is powered on and any IP interface is reachable to the network. Because of this capability, the loopback address is the best way to manage the switch in-band. Layer 3 process and features are also bound to the loopback interface to ensure process resiliency. The loopback address is commonly a host address with a 32-bit address mask. Allocate the loopback address from a unique network range that is not part of any other internal network summary range. interface Loopback 0 ip address [ip address] 255.255.255.255 ip pim sparse-mode Step 11: Bind the device processes for SNMP, SSH, PIM, TACACS+ and NTP to the loopback interface address for optimal resiliency: snmp-server trap-source Loopback0 ip ssh source-interface Loopback0 ip pim register-source Loopback0 ip tacacs source-interface Loopback0 ntp source Loopback0 Step 12: Configure IP Multicast routing. IP Multicast allows a single IP data stream to be replicated by the infrastructure (routers and switches) and sent from a single source to multiple receivers. Using IP Multicast is much more efficient than multiple individual unicast streams or a Broadcast stream that would propagate everywhere. IP Telephony MOH and IP Video Broadcast Streaming are two examples of IP Multicast applications. To receive a particular IP Multicast data stream, end hosts must join a multicast group by sending an IGMP message to their local multicast router. In a traditional IP Multicast design, the local router consults another router in the network that is acting as an RP to map the receivers to active sources so they can join their streams. In this design, which is based on sparse mode multicast operation, Auto RP is used to provide a simple yet scalable way to provide a highly resilient RP environment. Enable IP Multicast routing on the platforms in the global configuration mode. ip multicast-routing 37
Step 13: Configure every Layer 3 switch and router to discover the IP Multicast RP with autorp. Use the ip pim autorp listener command to allow for discovery across sparse mode links. This configuration provides for future scaling and control of the IP Multicast environment and can change based on network needs and design. ip pim autorp listener Step 14: Enable sparse mode multicast operation for all Layer 3 interfaces in the network. ip pim sparse-mode Procedure 2 Connect to the MPLS PE Router Step 1: Assign the interface bandwidth. The bandwidth value should correspond to the actual interface speed. Or, if you are using a subrate service, you should use the policed rate from the carrier. The example shows a Gigabit interface (1000 Mbps) with a subrate of 10 Mbps. interface [interface type] [number] bandwidth [bandwidth (kbps)] Tech Tip Command Reference: bandwidth kbps 10 Mbps = 10,000 kbps Step 2: Assign the IP address and netmask of the WAN interface. The IP addressing used between CE and PE routers must be negotiated with your MPLS carrier. Typically, you d use a point-to-point netmask of 255.255.255.252. interface [interface type] [number] ip address [IP address] [netmask] Step 3: Administratively enable the interface and disable CDP. The use of CDP on external interfaces is not recommended. interface [interface type] [number] no cdp enable no shutdown Example interface GigabitEthernet0/0 bandwidth 10000 ip address 192.168.3.9 255.255.255.252 no cdp enable no shutdown 38
Procedure 3 Configure WAN routing The remote-site CE routers are configured either for dynamic routing with BGP or are statically routed. If you have a remote-site design that includes sites with dual WAN links or you do not want to have your MPLS carrier make changes or modifications, use the BGP option. This is the recommended approach, and assumes that the WAN-aggregation CE router has already been configured for BGP. If your remote-site design only uses single WAN links and you don t anticipate adding or modifying IP networks at the remote site, then you can use the statically routed option. The MPLS carrier is responsible for configuring static IP routing within the MPLS network. Tech Tip If you do not use dynamic routing with BGP, then the MPLS carrier must configure a set of static routes on its PE routers for the WAN-aggregation site and for each of the remote sites. Site-specific routing details must be shared with your MPLS carrier. Option 1: BGP dynamic routing with MPLS carrier Step 1: Enable BGP. To complete this step, a BGP ASN is required. You might be able to reuse the same value used on the MPLS VPN CE from the WAN-aggregation site. Consult with your MPLS carrier on the requirements for the ASN. router bgp 65511 no synchronization bgp router-id [IP address of Loopback0] bgp log-neighbor-changes no auto-summary Step 2: Configure ebgp. Configure BGP with the MPLS carrier PE device. The MPLS carrier must provide their ASN (the ASN in the previous step is the ASN identifying your site). Because the carrier PE router uses a different ASN, this configuration is considered an external BGP (ebgp) connection. The CE router advertises only network routes to the PE via BGP in the following cases: The route is specified in network statements and is present in the local routing table. The route is redistributed into BGP (not applicable in the remote-site use case). It is desirable to advertise a route for the PE-CE link, so you should include this network in a network statement. You can use this to determine router reachability, for troubleshooting. Similarly, you must configure BGP to advertise the loopback network for the router. 39
You must advertise the remote-site LAN networks. The IP assignment for the remote sites was designed so that all of the networks in use can be summarized within a single aggregate route. The aggregate address configured below suppresses the more specific routes. If any LAN network is present in the route table, the aggregate is advertised to the MPLS PE, which offers a measure of resiliency. If the various LAN networks cannot be summarized, you must list each individually. You must add a separate network statement for the loopback address. router bgp 65511 network [PE-CE link network] mask [PE-CE link netmask] network [Loopback network] mask 255.255.255.255 network [DATA network] mask [netmask] network [VOICE network] mask [netmask] aggregate-address [summary IP address] [summary netmask] summary-only neighbor [IP address of PE] remote-as [carrier ASN] Example router bgp 65511 no synchronization bgp router-id 10.255.251.206 bgp log-neighbor-changes network 192.168.3.8 mask 255.255.255.252 network 10.255.251.206 mask 255.255.255.255 network 10.5.12.0 mask 255.255.255.0 network 10.5.13.0 mask 255.255.255.0 aggregate-address 10.5.8.0 255.255.248.0 summary-only neighbor 192.168.3.10 remote-as 65401 no auto-summary Option 2: Static routing with service provider This option has remote sites using static routing to the MPLS WAN to forward all traffic to the WAN-aggregation site. Step 1: Enter a default route for traffic forwarded to the WAN-aggregation site. ip route 0.0.0.0 0.0.0.0 192.168.3.10 Step 2: For the MPLS carrier for each remote site, provide the remote-site specific IP range and the chosen loopback IP address for the router. This properly configures the static routes to the remote site. Tech Tip For each remote site with static routing, the WAN-aggregation CE router must have a corresponding static host route for that site s loopback address. 40
Procedure 4 Connect router to access-layer switch Reader Tip This guide includes only the additional steps to complete the distribution-layer configuration. For complete access-layer configuration details, see the Campus Wired LAN Technology Design Guide. If you are using a remote-site distribution layer, skip to the Deploying a WAN Remote-Site Distribution Layer chapter of this guide. Layer 2 EtherChannels are used to interconnect the CE router to the access layer in the most resilient method possible. If your access-layer device is a single, fixed-configuration switch, a simple Layer 2 trunk between the router and switch is used. In the access-layer design, the remote sites use collapsed routing, with 802.1Q trunk interfaces to the LAN access layer. The VLAN numbering is locally significant only. Option 1: Layer 2 EtherChannel from router to access-layer switch Step 1: Configure port-channel interface on the router. interface Port-channel1 description EtherChannel link to RS206-A2960S no shutdown Step 2: Configure EtherChannel member interfaces on the router. Configure the physical interfaces to tie to the logical port-channel by using the channel-group command. The number for the port-channel and channel-group must match. Not all router platforms can support LACP to negotiate with the switch, so you configure EtherChannel statically. interface GigabitEthernet0/1 description RS206-A2960S Gig1/0/24 interface GigabitEthernet0/2 description RS206-A2960S Gig2/0/24 interface range GigabitEthernet0/1, GigabitEthernet0/2 no ip address channel-group 1 no shutdown Step 3: Configure EtherChannel member interfaces on the access-layer switch. Connect the router EtherChannel uplinks to separate switches in the access layer switch stack, or in the case of the Cisco Catalyst 4507R+E distribution layer, to separate redundant modules for additional resiliency. The physical interfaces that are members of a Layer 2 EtherChannel are configured prior to configuring the logical port-channel interface. Doing the configuration in this order allows for minimal configuration and reduces errors because most of the commands entered to a port-channel interface are copied to its members interfaces and do not require manual replication. 41
Configure two physical interfaces to be members of the EtherChannel. Also, apply the egress QoS macro that was defined in the LAN switch platform configuration procedure to ensure traffic is prioritized appropriately. Not all connected router platforms can support LACP to negotiate with the switch, so you configure EtherChannel statically. interface GigabitEthernet1/0/24 description Link to RS206-3925-1 Gig0/1 interface GigabitEthernet2/0/24 description Link to RS206-3925-1 Gig0/2 interface range GigabitEthernet1/0/24, GigabitEthernet2/0/24 switchport macro apply EgressQoS channel-group 1 mode on logging event link-status logging event trunk-status logging event bundle-status Step 4: Configure EtherChannel trunk on the access-layer switch. Use an 802.1Q trunk for the connection, which allows the router to provide the Layer 3 services to all the VLANs defined on the access-layer switch. Prune the VLANs allowed on the trunk to only the VLANs that are active on the access-layer switch. When using EtherChannel, the interface type is port-channel, and the number must match the channel group configured in Step 3. Set DHCP Snooping and Address Resolution Protocol (ARP) inspection to trust. interface Port-channel1 description EtherChannel link to RS206-3925-1 switchport trunk encapsulation dot1q switchport trunk allowed vlan 64,69 switchport mode trunk ip arp inspection trust spanning-tree portfast trunk ip dhcp snooping trust no shutdown The Cisco Catalyst 2960-S Series and 4500 Series switches do not require the switchport trunk encapsulation dot1q command. Option 2: Layer 2 trunk from router to access-layer switch Step 1: Enable the physical interface on the router. interface GigabitEthernet0/2 description RS202-A3560X Gig1/0/24 no ip address no shutdown 42
Step 2: Configure the trunk on the access-layer switch. Use an 802.1Q trunk for the connection, which allows the router to provide the Layer 3 services to all the VLANs defined on the access-layer switch. Prune the VLANs allowed on the trunk to only the VLANs that are active on the access-layer switch. Set DHCP Snooping and Address Resolution Protocol (ARP) inspection to trust. interface GigabitEthernet1/0/24 description Link to RS201-2911 Gig0/2 switchport trunk encapsulation dot1q switchport trunk allowed vlan 64,69 switchport mode trunk ip arp inspection trust spanning-tree portfast trunk macro apply EgressQoS logging event link-status logging event trunk-status ip dhcp snooping trust no shutdown The Cisco Catalyst 2960-S Series and 4500 Series switches do not require the switchport trunk encapsulation dot1q command. Option 3: Layer 2 trunk from Cisco 881 router to access-layer switch This option uses a single Ethernet connection in order to connect the remote-site Cisco 881 Integrated Services Router to a single-member access switch. This option differs significantly from the previous options because an embedded Ethernet switch provides the LAN connectivity of the Cisco 881 router. Step 1: Configure necessary VLANs on the embedded switch of the remote-site router. vlan 64 name Wired-Data vlan 69 name Wired-Voice vlan 999 name Native Step 2: Configure the remote-site router s connection to the remote-site Ethernet switch. interface FastEthernet0 switchport trunk native vlan 999 switchport trunk allowed vlan 1,2,64,69,1002-1005 switchport mode trunk no ip address no shutdown Tech Tip The embedded switch on the Cisco 881 Integrated Services Router requires that the default VLANs be allowed on trunks. To maintain security and configuration consistency, these VLANs (1, 2, 1002, 1003, 1004, and 1005) are pruned on the access-switch side of the trunk. 43
Step 3: Configure the trunk on the access-layer switch. Use an 802.1Q trunk for the connection. This allows the router to provide the Layer 3 services to all the VLANs defined on the access-layer switch. Prune the VLANs allowed on the trunk to only the VLANs that are active on the access switch, and then set DHCP Snooping and Address Resolution Protocol (ARP) inspection to trust. interface GigabitEthernet1/0/24 description Link to RS4-881 Fast0 switchport trunk native vlan 999 switchport trunk encapsulation dot1q switchport trunk allowed vlan 64,69 switchport mode trunk ip arp inspection trust macro apply EgressQoS logging event link-status logging event trunk-status ip dhcp snooping trust no shutdown The Cisco Catalyst 2960-S Series and 4500 Series switches do not require the switchport trunk encapsulation dot1q command. Procedure 5 Configure access-layer routing Option 1: Layer 2 EtherChannel or Layer 2 trunk Step 1: Create subinterfaces and assign VLAN tags. After you have enabled the physical interface or port-channel, you can map the appropriate data or voice subinterfaces to the VLANs on the LAN switch. The subinterface number does not need to equate to the 802.1Q tag, but making them the same simplifies the overall configuration. The subinterface portion of the configuration should be repeated for all data or voice VLANs. interface [type][number].[sub-interface number] encapsulation dot1q [dot1q VLAN tag] Step 2: Configure IP settings for each subinterface. This design uses an IP addressing convention with the default gateway router assigned an IP address and IP mask combination of N.N.N.1 255.255.255.0 where N.N.N is the IP network and 1 is the IP host. When using a centralized DHCP server, routers with LAN interfaces connected to a LAN using DHCP for endstation IP addressing must use an IP helper. This is the preferred method. An alternate option for local DHCP server configuration is shown in the following procedure. If the remote-site router is the first router of a dual-router design, then HSRP is configured at the access layer. This requires a modified IP configuration on each subinterface. interface [type][number].[sub-interface number] description [usage] ip address [LAN network 1] [LAN network 1 netmask] ip helper-address 10.4.48.10 ip pim sparse-mode 44
Option 2: Layer 2 trunk from Cisco 881 router to access-layer switch Step 1: Configure IP settings for each subinterface. This design uses an IP addressing convention with the default gateway router assigned an IP address and IP mask combination of N.N.N.1 255.255.255.0 where N.N.N is the IP network and 1 is the IP host. When using a centralized DHCP server, routers with LAN interfaces connected to a LAN using DHCP for endstation IP addressing must use an IP helper. This is the preferred method. An alternate option for local DHCP server configuration is shown in the following procedure. interface [VLAN number] description [usage] ip address [LAN network 1] [LAN network 1 netmask] ip helper-address 10.4.48.10 ip pim sparse-mode Example: Layer 2 EtherChannel interface Port-channel1 no ip address no shutdown interface Port-channel1.64 description Data encapsulation dot1q 64 ip address 10.5.12.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode interface Port-channel1.69 description Voice encapsulation dot1q 69 ip address 10.5.13.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode 45
Example: Layer 2 Link interface GigabitEthernet0/2 no ip address no shutdown interface GigabitEthernet0/2.64 description Data encapsulation dot1q 64 ip address 10.5.12.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode interface GigabitEthernet0/2.69 description Voice encapsulation dot1q 69 ip address 10.5.13.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode Example: Layer 2 Link from Cisco 881 ISR interface Vlan64 description Data ip address 10.5.28.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode interface Vlan69 description Voice ip address 10.5.29.1 255.255.255.0 ip helper-address 10.4.48.10 ip pim sparse-mode Procedure 6 Configure remote-site DHCP (Optional) The previous procedure assumes the DHCP service has been configured centrally and uses the ip helperaddress command to forward DHCP requests to the centralized DHCP server. If you choose to run a local DHCP server on the remote-site router instead of centralizing the DHCP service, complete this procedure. This procedure uses a local DHCP service on the router in order to assign basic network configuration for IP phones, wireless access points, users laptop and desktop computers, and other endpoint devices. 46
Tech Tip If you intend to use a dual-router remote-site design, you should use a resilient DHCP solution, such as a centralized DHCP server. Options for resilient DHCP at the remotesite include using IOS on a distribution-layer switch stack or implementing a dedicated DHCP server solution. Step 1: Remove the previously configured ip helper-address commands for any interface that uses a local DHCP server. Step 2: Configure a DHCP scope for data endpoints, excluding DHCP assignment for the first 19 addresses in the subnet. ip dhcp excluded-address 10.5.4.1 10.11.4.19 ip dhcp pool DHCP-Wired-Data network 10.5.4.0 255.255.255.0 default-router 10.5.4.1 domain-name cisco.local dns-server 10.4.48.10 Step 3: Configure a DHCP scope for voice endpoints, excluding DHCP assignment for the first 19 addresses in the subnet. Step 4: Voice endpoints require an option field to tell them where to find their initial configuration. Different vendors use different option fields, so the number may vary based on the voice product you choose (for example, Cisco uses DHCP option 150). ip dhcp excluded-address 10.5.5.1 10.11.5.19 ip dhcp pool DHCP-Wired-Voice network 10.5.5.0 255.255.255.0 default-router 10.5.5.1 domain-name cisco.local dns-server 10.4.48.10 Procedure 7 through Procedure 11 are only relevant for the dual-router design. Procedure 7 Configure access-layer HSRP If you are using a dual-router design, complete this procedure. You need to configure HSRP in order to enable the use of a virtual IP (VIP) address as a default gateway that is shared between two routers. The HSRP active router is the MPLS CE router connected to the primary MPLS carrier, and the HSRP standby router is the router connected to the secondary MPLS carrier or backup link. 47
In this procedure, you configure the HSRP active router with a standby priority that is higher than the HSRP standby router. The router with the higher standby priority value is elected as the HSRP active router. The preempt option allows a router with a higher priority to become the HSRP active, without waiting for a scenario where there is no router in the HSRP active state. The relevant HSRP parameters for the router configuration are shown in the following table. Table 11 - WAN remote-site HSRP parameters (dual-router design) Router HSRP role Virtual IP address (VIP) Real IP address HSRP priority PIM DR priority MPLS CE (primary) Active.1.2 110 110 MPLS CE (secondary) or DMVPN spoke Standby.1.3 105 105 The assigned IP addresses override those configured in the previous procedure, so the default gateway IP address remains consistent across locations with single or dual routers. The dual-router access-layer design requires a modification for resilient multicast. The PIM designated router (DR) should be on the HSRP active router. The DR is normally elected based on the highest IP address, and it has no awareness of the HSRP configuration. In this design, assigning the HSRP active router a lower real IP address than the HSRP standby router requires a modification to the PIM configuration. You can influence the PIM DR election by explicitly setting the DR priority on the LAN-facing subinterfaces for the routers. Tech Tip The HSRP priority and PIM DR priority are shown in the previous table to be the same value; however, you are not required to use identical values. Step 1: Configure HSRP. interface [type][number].[sub-interface number] ip address [LAN network 1 address] [LAN network 1 netmask] ip pim dr-priority 110 standby version 2 standby 1 ip [LAN network 1 gateway address] standby 1 priority 110 standby 1 preempt standby 1 authentication md5 key-string c1sco123 48
Step 2: Repeat this procedure for all data or voice subinterfaces. Example: Layer 2 link interface GigabitEthernet0/2 no ip address no shutdown interface GigabitEthernet0/2.64 description Data encapsulation dot1q 64 ip address 10.5.12.2 255.255.255.0 ip helper-address 10.4.48.10 ip pim dr-priority 110 ip pim sparse-mode standby version 2 standby 1 ip 10.5.12.1 standby 1 priority 110 standby 1 preempt standby 1 authentication md5 key-string c1sco123 interface GigabitEthernet0/2.69 description Voice encapsulation dot1q 69 ip address 10.5.13.2 255.255.255.0 ip helper-address 10.4.48.10 ip pim dr-priority 110 ip pim sparse-mode standby version 2 standby 1 ip 10.5.13.1 standby 1 priority 110 standby 1 preempt standby 1 authentication md5 key-string c1sco123 Procedure 8 Configure the transit network If you are using a dual-router design, complete this procedure. The transit network is configured between the two routers. This network is used for router-router communication and to avoid hairpinning. The transit network should use an additional subinterface on the router interface that is already being used for data or voice. There are no end stations connected to this network, so HSRP and DHCP are not required. Step 1: On the primary MPLS CE router, configure the transit network interface. interface [type][number].[sub-interface number] encapsulation dot1q [dot1q VLAN tag] ip address [transit net address] [transit net netmask] ip pim sparse-mode 49
Example interface GigabitEthernet0/2.99 description Transit Net encapsulation dot1q 99 ip address 10.5.8.1 255.255.255.252 ip pim sparse-mode Step 2: On the access-layer switch, add the transit network VLAN. vlan 99 name Transit-net Step 3: Add the transit network VLAN to the existing access-layer switch trunk. interface GigabitEthernet1/0/24 switchport trunk allowed vlan add 99 Procedure 9 Configure EIGRP (LAN side) If you are using a dual-router design, complete this procedure. You must configure a routing protocol between the two routers. This ensures that the HSRP active router has full reachability information for all WAN remote sites. Step 1: Enable EIGRP-100 facing the access layer. In this design, all LAN-facing interfaces and the loopback must be EIGRP interfaces. All interfaces except the transit-network subinterface should remain passive. The network range must include all interface IP addresses either in a single network statement or in multiple network statements. This design uses a best practice of assigning the router ID to a loopback address. Do not include the WAN interface (MPLS PE-CE link interface) as an EIGRP interface. router eigrp 100 network [network] [inverse mask] passive-interface default no passive-interface [Transit interface] eigrp router-id [IP address of Loopback0] no auto-summary Step 2: Redistribute BGP into EIGRP-100. A default metric redistributes the BGP routes into EIGRP. By default, only the WAN bandwidth and delay values are used for metric calculation. router eigrp 100 default-metric [WAN bandwidth] [WAN delay] 255 1 1500 redistribute bgp 65511 50
Tech Tip Command Reference: default-metric bandwidth delay reliability loading mtu bandwidth Minimum bandwidth of the route in kilobytes per second delay Route delay in tens of microseconds. Example router eigrp 100 default-metric 100000 100 255 1 1500 network 10.4.0.0 0.1.255.255 network 10.255.0.0 0.0.255.255 redistribute bgp 65511 passive-interface default no passive-interface GigabitEthernet0/2.99 eigrp router-id 10.255.251.206 no auto-summary Procedure 10 Configure BGP If you are using a dual-router design, complete this procedure. Step 1: On both remote-site MPLS CE routers, configure ibgp and enable the next-hop-self configuration option. The dual-carrier MPLS design requires that a BGP link is configured between the CE routers. Because the CE routers are using the same ASN, this configuration is considered an internal BGP (ibgp) connection. This design uses ibgp peering using the transit network, which requires the next-hop-self-configuration option. You must complete this step on both remote-site MPLS CE routers. Note, the ibgp session will not be established until you complete the transit network and EIGRP (LAN-side) steps. router bgp 65511 neighbor [ibgp neighbor Transit Net IP] remote-as 65511 neighbor [ibgp neighbor Transit Net IP] next-hop-self 51
Step 2: Configure BGP to prevent the remote site from becoming a transit AS. By default, BGP readvertises all BGP-learned routes. In the dual-mpls design, this means that MPLS-A routes will be advertised to MPLS-B and vice-versa. In certain cases, when a link to a MPLS hub has failed, remote sites will advertise themselves as a transit autonomous system, providing access between the two carriers. Unless the remote site has been specifically designed for this type of routing behavior, with a high bandwidth connection, it is a best practice to disable the site from becoming a transit site. You must use a route-map and an as-path access-list filter. You need to apply this route-map on both remote-site MPLS CE routers. Each router will apply this outbound to the neighbor for its respective MPLS carrier. router bgp 65511 neighbor [IP address of PE] route-map NO-TRANSIT-AS out ip as-path access-list 10 permit ^$ route-map NO-TRANSIT-AS permit 10 match as-path 10 Tech Tip The regular expression ^$ corresponds to routes originated from the remote-site. This type of filter allows for only the locally originated routes to be advertised. Step 3: Tune BGP routing to prefer the primary MPLS carrier. BGP uses a well-known rule set in order to determine the best path when the same IP route prefix is reachable via two different paths. The MPLS dual-carrier design in many cases provides two equal cost paths, and it is likely that the first path selected will remain the active path unless the routing protocol detects a failure. Accomplishing the design goal of deterministic routing and primary/secondary routing behavior necessitates tuning BGP. This requires the use of a route-map and an as-path access-list filter. router bgp 65511 neighbor [IP address of PE] route-map PREFER-MPLS-A in Step 4: Apply a route-map inbound to the neighbor for the primary MPLS carrier only. ip as-path access-list 1 permit _65401$ route-map PREFER-MPLS-A permit 10 match as-path 1 set local-preference 200 route-map PREFER-MPLS-A permit 20 52
Tech Tip The regular expression _65401$ corresponds to routes originated from the AS 65401 (MPLS-A). This allows BGP to selectively modify the routing information for routes originated from this AS. In this example, the BGP local preference is 200 for the primary MPLS carrier. Routes originated from the secondary MPLS carrier continue to use their default local preference of 100. Step 5: Add a loopback network for the secondary router. router bgp 65511 network [Secondary router loopback network] mask 255.255.255.255 Procedure 11 Enable Enhanced Object Tracking If you are using a dual-router design, complete this procedure. The HSRP active router remains the active router unless the router is reloaded or fails. Having the HSRP router remain as the active router can lead to undesired behavior. If the primary MPLS VPN transport were to fail, the HSRP active router would learn an alternate path through the transit network to the HSRP standby router and begin to forward traffic across the alternate path. This is sub-optimal routing, and you can address it by using EOT. The HSRP active router (primary MPLS CE) can use the IP SLA feature to send echo probes to its MPLS PE router, and if the PE router becomes unreachable, then the router can lower its HSRP priority, so that the HSRP standby router can preempt and become the HSRP active router. This procedure is valid only on the router connected to the primary transport (MPLS VPN). Step 1: Enable the IP SLA probe, and then send standard ICMP echo (ping) probes at 15-second intervals. Responses must be received before the timeout of 1000 ms expires. If you are using the MPLS PE router as the probe destination, the destination address is the same as the BGP neighbor address configured in Procedure 3. ip sla 100 icmp-echo [probe destination IP address] source-interface [WAN interface] threshold 1000 timeout 1000 frequency 15 ip sla schedule 100 life forever start-time now Step 2: Configure EOT based on the IP SLA probe. The object being tracked is the reachability success or failure of the probe. If the probe is successful, the tracked object status is Up; if it fails, the tracked object status is Down. track 50 ip sla 100 reachability 53
Step 3: Link HSRP with the tracked object. All data or voice subinterfaces should enable HSRP tracking. HSRP can monitor the tracked object status. If the status is down, the HSRP priority is decremented by the configured priority. If the decrease is large enough, the HSRP standby router preempts. interface [interface type] [number].[sub-interface number] standby 1 track 50 decrement 10 Example interface GigabitEthernet 0/2.64 standby 1 track 50 decrement 10 interface GigabitEthernet 0/2.69 standby 1 track 50 decrement 10 track 50 ip sla 100 reachability ip sla 100 icmp-echo 192.168.3.10 source-interface GigabitEthernet0/0 timeout 1000 threshold 1000 frequency 15 ip sla schedule 100 life forever start-time now PROCESS Adding a Secondary MPLS Link on an Existing MPLS CE Router 1. Connect to MPLS PE router 2. Configure BGP for dual-link design This process includes the additional steps necessary to complete the configuration of an MPLS CE router for an MPLS WAN dual-carrier remote site (single-router, dual-link). The following procedures assume that the configuration of an MPLS CE router for an MPLS WAN remote site (single-router, single-link) has already been completed and BGP dynamic routing has been configured. Only the additional procedures to add an additional MPLS link to the running MPLS CE router are included here. 54
The following figure provides details on how to add a second MPLS backup link on an existing remote-site MPLS CE router. Figure 15 - Flowchart for adding an MPLS backup configuration MPLS CE Router Configuration Complete Adding Second MPLS Link on Existing MPLS CE Router Configuration Procedures YES Add MPLS Secondary Link? 1. Connect to MPLS PE Router 2. Configure BGP for Dual-linklink NO Site Complete 2130 Procedure 1 Connect to MPLS PE router This procedure applies to the interface used to connect the secondary or additional MPLS carrier. Step 1: Assign an interface bandwidth value that corresponds to the actual interface speed. If you are using a subrate service, use the policed rate from the carrier. The example shows a Gigabit interface (1000 Mbps) with a subrate of 10 Mbps. interface [interface type] [number] bandwidth [bandwidth (kbps)] 55